summaryrefslogtreecommitdiff
path: root/fs/xfs/scrub/parent.c
blob: ec351a393b2d5651bb9f24cad94d1c5592962b9a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
// SPDX-License-Identifier: GPL-2.0+
/*
 * Copyright (C) 2017 Oracle.  All Rights Reserved.
 * Author: Darrick J. Wong <darrick.wong@oracle.com>
 */
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_shared.h"
#include "xfs_format.h"
#include "xfs_trans_resv.h"
#include "xfs_mount.h"
#include "xfs_log_format.h"
#include "xfs_trans.h"
#include "xfs_inode.h"
#include "xfs_icache.h"
#include "xfs_dir2.h"
#include "xfs_dir2_priv.h"
#include "scrub/scrub.h"
#include "scrub/common.h"
#include "scrub/parent.h"
#include "scrub/repair.h"

/* Set us up to scrub parents. */
int
xchk_setup_parent(
	struct xfs_scrub	*sc)
{
	int			error;

	if (sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) {
		error = xrep_setup_parent(sc);
		if (error)
			return error;
	}

	return xchk_setup_inode_contents(sc, 0);
}

/* Parent pointers */

/* Look for an entry in a parent pointing to this inode. */

struct xchk_parent_ctx {
	struct dir_context	dc;
	struct xfs_scrub	*sc;
	xfs_ino_t		ino;
	xfs_nlink_t		nlink;
	bool			cancelled;
};

/* Look for a single entry in a directory pointing to an inode. */
STATIC int
xchk_parent_actor(
	struct dir_context	*dc,
	const char		*name,
	int			namelen,
	loff_t			pos,
	u64			ino,
	unsigned		type)
{
	struct xchk_parent_ctx	*spc;
	int			error = 0;

	spc = container_of(dc, struct xchk_parent_ctx, dc);
	if (spc->ino == ino)
		spc->nlink++;

	/*
	 * If we're facing a fatal signal, bail out.  Store the cancellation
	 * status separately because the VFS readdir code squashes error codes
	 * into short directory reads.
	 */
	if (xchk_should_terminate(spc->sc, &error))
		spc->cancelled = true;

	return error;
}

/* Count the number of dentries in the parent dir that point to this inode. */
STATIC int
xchk_parent_count_parent_dentries(
	struct xfs_scrub	*sc,
	struct xfs_inode	*parent,
	xfs_nlink_t		*nlink)
{
	struct xchk_parent_ctx	spc = {
		.dc.actor	= xchk_parent_actor,
		.ino		= sc->ip->i_ino,
		.sc		= sc,
	};
	size_t			bufsize;
	loff_t			oldpos;
	uint			lock_mode;
	int			error = 0;

	/*
	 * If there are any blocks, read-ahead block 0 as we're almost
	 * certain to have the next operation be a read there.  This is
	 * how we guarantee that the parent's extent map has been loaded,
	 * if there is one.
	 */
	lock_mode = xfs_ilock_data_map_shared(parent);
	if (parent->i_df.if_nextents > 0)
		error = xfs_dir3_data_readahead(parent, 0, 0);
	xfs_iunlock(parent, lock_mode);
	if (error)
		return error;

	/*
	 * Iterate the parent dir to confirm that there is
	 * exactly one entry pointing back to the inode being
	 * scanned.
	 */
	bufsize = (size_t)min_t(loff_t, XFS_READDIR_BUFSIZE,
			parent->i_disk_size);
	oldpos = 0;
	while (true) {
		error = xfs_readdir(sc->tp, parent, &spc.dc, bufsize);
		if (error)
			goto out;
		if (spc.cancelled) {
			error = -EAGAIN;
			goto out;
		}
		if (oldpos == spc.dc.pos)
			break;
		oldpos = spc.dc.pos;
	}
	*nlink = spc.nlink;
out:
	return error;
}

/*
 * Try to iolock the parent dir @dp in shared mode and the child dir @sc->ip
 * exclusively.
 */
int
xchk_parent_lock_two_dirs(
	struct xfs_scrub	*sc,
	struct xfs_inode	*dp)
{
	int			error = 0;

	/* Callers shouldn't do this, but protect ourselves anyway. */
	if (dp == sc->ip) {
		ASSERT(dp != sc->ip);
		return -EDEADLOCK;
	}

	xchk_iunlock(sc, sc->ilock_flags);
	while (true) {
		if (xchk_should_terminate(sc, &error))
			return error;
		xfs_ilock(dp, XFS_IOLOCK_SHARED);
		if (xchk_ilock_nowait(sc, XFS_IOLOCK_EXCL))
			break;
		xfs_iunlock(dp, XFS_IOLOCK_SHARED);
	}

	return 0;
}

/*
 * Given the inode number of the alleged parent of the inode being
 * scrubbed, try to validate that the parent has exactly one directory
 * entry pointing back to the inode being scrubbed.
 */
STATIC int
xchk_parent_validate(
	struct xfs_scrub	*sc,
	xfs_ino_t		parent_ino)
{
	struct xfs_mount	*mp = sc->mp;
	struct xfs_inode	*dp = NULL;
	xfs_nlink_t		expected_nlink;
	xfs_nlink_t		nlink;
	int			error = 0;

	if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
		return 0;

	/* '..' must not point to ourselves. */
	if (sc->ip->i_ino == parent_ino) {
		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		return 0;
	}

	/*
	 * If we're an unlinked directory, the parent /won't/ have a link
	 * to us.  Otherwise, it should have one link.
	 */
	expected_nlink = VFS_I(sc->ip)->i_nlink == 0 ? 0 : 1;

	/*
	 * Grab the parent directory inode.  This must be released before we
	 * cancel the scrub transaction.
	 *
	 * If _iget returns -EINVAL or -ENOENT then the parent inode number is
	 * garbage and the directory is corrupt.  If the _iget returns
	 * -EFSCORRUPTED or -EFSBADCRC then the parent is corrupt which is a
	 *  cross referencing error.  Any other error is an operational error.
	 */
	error = xfs_iget(mp, sc->tp, parent_ino, XFS_IGET_UNTRUSTED, 0, &dp);
	if (error == -EINVAL || error == -ENOENT) {
		error = -EFSCORRUPTED;
		xchk_fblock_process_error(sc, XFS_DATA_FORK, 0, &error);
		return error;
	}
	if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error))
		return error;
	if (dp == sc->ip || !S_ISDIR(VFS_I(dp)->i_mode)) {
		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		goto out_rele;
	}

	/*
	 * Metadata and regular inodes cannot cross trees.  This property
	 * cannot change without a full inode free and realloc cycle, so it's
	 * safe to check this without holding locks.
	 */
	if (xfs_is_metadata_inode(dp) ^ xfs_is_metadata_inode(sc->ip)) {
		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		goto out_rele;
	}

	/*
	 * We prefer to keep the inode locked while we lock and search its
	 * alleged parent for a forward reference.  If we can grab the iolock
	 * of the alleged parent, then we can move ahead to counting dirents
	 * and checking nlinks.
	 *
	 * However, if we fail to iolock the alleged parent while holding the
	 * child iolock, we have no way to tell if a blocking lock() would
	 * result in an ABBA deadlock.  Release the lock on the child, then
	 * try to lock the alleged parent and trylock the child.
	 */
	if (!xfs_ilock_nowait(dp, XFS_IOLOCK_SHARED)) {
		error = xchk_parent_lock_two_dirs(sc, dp);
		if (error)
			goto out_rele;

		error = xfs_dir_lookup(sc->tp, sc->ip, &xfs_name_dotdot,
				&parent_ino, NULL);
		if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, 0, &error))
			goto out_unlock;

		/*
		 * After relocking the child directory, the '..' entry points
		 * to a different parent than before.  This means someone moved
		 * the child elsewhere in the directory tree, which means that
		 * the parent link is now correct and we're done.
		 */
		if (parent_ino != dp->i_ino)
			goto out_unlock;
	}

	/* Look for a directory entry in the parent pointing to the child. */
	error = xchk_parent_count_parent_dentries(sc, dp, &nlink);
	if (!xchk_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error))
		goto out_unlock;

	/*
	 * Ensure that the parent has as many links to the child as the child
	 * thinks it has to the parent.
	 */
	if (nlink != expected_nlink)
		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);

out_unlock:
	xfs_iunlock(dp, XFS_IOLOCK_SHARED);
out_rele:
	xchk_irele(sc, dp);
	return error;
}

/* Scrub a parent pointer. */
int
xchk_parent(
	struct xfs_scrub	*sc)
{
	struct xfs_mount	*mp = sc->mp;
	xfs_ino_t		parent_ino;
	int			error;

	/*
	 * If we're a directory, check that the '..' link points up to
	 * a directory that has one entry pointing to us.
	 */
	if (!S_ISDIR(VFS_I(sc->ip)->i_mode))
		return -ENOENT;

	/* We're not a special inode, are we? */
	if (!xfs_verify_dir_ino(mp, sc->ip->i_ino)) {
		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		return 0;
	}

	/*
	 * The VFS grabs a read or write lock via i_rwsem before it reads
	 * or writes to a directory.  If we've gotten this far we've
	 * already obtained IOLOCK_EXCL, which (since 4.10) is the same as
	 * getting a write lock on i_rwsem.  Therefore, it is safe for us
	 * to drop the ILOCK here in order to do directory lookups.
	 */
	xchk_iunlock(sc, XFS_ILOCK_EXCL | XFS_MMAPLOCK_EXCL);

	/* Look up '..' */
	error = xfs_dir_lookup(sc->tp, sc->ip, &xfs_name_dotdot, &parent_ino,
			NULL);
	if (!xchk_fblock_process_error(sc, XFS_DATA_FORK, 0, &error))
		return error;
	if (!xfs_verify_dir_ino(mp, parent_ino)) {
		xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		return 0;
	}

	/* Is this the root dir?  Then '..' must point to itself. */
	if (sc->ip == mp->m_rootip) {
		if (sc->ip->i_ino != mp->m_sb.sb_rootino ||
		    sc->ip->i_ino != parent_ino)
			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		return 0;
	}

	/* Is this the metadata root dir?  Then '..' must point to itself. */
	if (sc->ip == mp->m_metadirip) {
		if (sc->ip->i_ino != mp->m_sb.sb_metadirino ||
		    sc->ip->i_ino != parent_ino)
			xchk_fblock_set_corrupt(sc, XFS_DATA_FORK, 0);
		return 0;
	}

	return xchk_parent_validate(sc, parent_ino);
}